skills.rest
adjohn
adjohnCOMMUNITY
Explore all skills by this author →
GitHub

why

Investigate design rationale behind code by querying seven evidence sources in parallel.

3|2|Updated Aug 28, 2026
Search Tags:#investigation#mcp#design rationale#git history#postmortem#evidence synthesis#code archaeology
One-click install
npx skills add https://github.com/adjohn/pstack --skill why-adjohn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: why
Source: https://github.com/adjohn/pstack/tree/main/skills/why
Command: npx skills add https://github.com/adjohn/pstack --skill why-adjohn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Code tells you what it does but rarely why it exists. This Skill reconstructs the historical motivation behind code—design decisions, tradeoffs, rejected alternatives, and business constraints—by searching fragmented evidence across source control, tickets, docs, chat, observability, error tracking, and analytics systems. ## Core Features & Use Cases - Parallel Multi-Source Investigation: Discovers available MCP servers at runtime, maps them to seven evidence categories, and spawns one investigator subagent per category to search concurrently. - Confidence-Calibrated Synthesis: A dedicated synthesizer merges findings into a cited narrative that strictly separates direct evidence, supported inferences, competing hypotheses, and explicit gaps. - Epistemic Discipline: Enforces a five-tier confidence framework with hedged phrasing rules, preventing confident-sounding guesses from masquerading as facts. - Use Case: Ask why a retry limit is set to 3 in a payment service. The Skill searches git history, Linear tickets, Notion docs, Slack threads, Datadog monitors, Sentry issues, and Databricks analytics, then returns a cited answer showing the threshold came from a postmortem action item. ## Quick Start Ask why the retry logic in backend/retry.ts was designed with exponential backoff and a limit of three attempts.

Frequently Asked Questions about why

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find out why code was written a certain way?▼

Anchor the investigation with git blame and git log on the target lines, then search PR discussions, issue trackers, design docs, chat threads, and observability data in parallel. This Skill automates that by spawning one investigator per evidence source and synthesizing cited findings.

What sources does this skill search for design rationale?▼

It covers seven categories: source control history via git and gh, issue trackers like Linear or Jira, long-form docs like Notion or Confluence, team chat like Slack, infrastructure observability like Datadog, error tracking like Sentry, and analytics warehouses like Databricks.

Does this skill require specific MCP servers to work?▼

Only source control via git and gh is guaranteed. The skill discovers available MCP servers at runtime, maps each to an evidence category, and searches whatever is available. Missing categories are documented as explicit gaps rather than silently skipped.

How does the skill handle conflicting evidence about a decision?▼

Contradictions are surfaced, not resolved. When a ticket and a PR description disagree, both are presented with citations so the user can judge. The synthesizer is explicitly forbidden from quietly picking the version that fits a tidier narrative.

When should I use the why skill versus reading the code directly?▼

Use it when you need motivation, not mechanics. Code shows what it does but not why it exists; rationale lives in commits, PRs, tickets, docs, and conversations. For understanding runtime behavior and mechanics, use the companion how skill instead.

Related Skills

Explore More Software Engineering Skills→
pckienuitpckienuit

codegraph

Find code symbols and trace dependency paths in a repository.

Community
Intermediate
nxtg-ainxtg-ai

repo-intake

Map a software repository's architecture, constraints, and key modules for implementation planning.

Official
Intermediate
atman-33atman-33

code-investigator

Investigate code to confirm specifications and generate evidence-based reports with file/line references.

Community
Advanced
rett-europerett-europe
2

reviewer-protocol

Enforce lockout semantics for reviewer rejections in coordination workflows.

Official
Advanced